Causal Confusion and Reward Misidentification in Preference-Based Reward Learning

Open in new window